What’s in this article
- What’s coming — GPT-5.5 plus Codex plus a built-in browser, in one session.
- Why “two years of research” matters — this isn’t an incremental bump.
- What designers should expect in their ChatGPT workflow.
- The Claude vs. ChatGPT decision if you’re picking one.
- How I’d actually use this when it drops.
I’m Mike Kwal. I keep one ChatGPT Plus seat alongside my Claude stack. So a model jump on the OpenAI side genuinely changes my workflow. Here’s what to expect.
What just happened
OpenAI’s next model — codenamed Spud internally — is expected to drop before April 30. Greg Brockman called it “two years of research.” That’s deliberately not “an iteration.” It’s framed as a leap.
The rumored package: GPT-5.5 + Codex + a built-in browser, all in one session. Describe an idea, the model writes the code, the browser deploys it. One conversation, start to finish. No tab switching.
For designers using ChatGPT today, this is the version that turns it from “a smarter chat” into “an actual co-builder.”
Why this matters for designers
Three shifts likely show up in your daily work.
The “single tool” promise gets real. Right now most designers run ChatGPT plus a code editor plus a deploy tool. GPT-5.5 in the rumored shape collapses those three into one window. Talk to the AI, see the code, see the deployed preview — all in the same conversation.
Copy work levels up. OpenAI has historically had a slight edge on creative writing. A two-years-of-research model widens that gap. For designers writing landing page copy, hero blocks, and ad creative, this is where ChatGPT starts pulling ahead of pure-Claude users on the writing layer.
Code work catches up. Claude has been the clear leader for code work since Opus 3.5. GPT-5.5 is rumored to close that gap. If it does, designers who pay for both ChatGPT Plus and Claude Pro stop having a clear winner — and start having a “use whichever fits the task” stack.
The risk for designers betting solely on Claude: missing the writing-layer lift that GPT-5.5 likely delivers. The risk of betting solely on ChatGPT: missing whatever Anthropic ships in response.
My $0.02 — How I’d actually use this
When Spud drops, here’s the playbook I’d run inside my studio.
Day 1: Side-by-side test on real client work. I’d take one current project — say a landing page rewrite — and give the same brief to both Claude and GPT-5.5. Same prompt, same context. Compare the outputs. The difference will tell me which tool wins this category for me right now. The same week, I’d run a code task on both — say “add an animated transition to this Webflow custom element” — and compare again.
Day 2: Update my daily prompts. If GPT-5.5 wins the writing tasks, my daily prompt library shifts. Hero copy, ad copy, blog drafts — those go to GPT. Code work and project-wide reasoning still go to Claude. I want each tool doing what it’s best at, not me defaulting to one out of habit.
Day 7: Decide if I keep both. $40/month for Claude + ChatGPT is justified if both are pulling weight. If GPT-5.5 makes ChatGPT noticeably better at the parts of my work where Claude is already strong (project-wide code, long context, security review), I’d reconsider whether I need both. My current bet is: keep both, because the cheap insurance of running two senior writers on every project is worth $20/month.
The signal I’m watching for: does the rumored “ChatGPT + Codex + browser” combo actually feel like one tool, or is it three apps in a trench coat? If it feels like one tool, that’s a meaningful shift in OpenAI’s UX, not just their model. If it feels stitched together, the model is what matters and the browser is a marketing feature.
The bigger lesson for designers: don’t fall in love with one AI tool. Fall in love with the outcome (great client work) and let the tool stack shift as the leaders shift.
Want the full playbook?
For the full Claude vs. ChatGPT vs. DeepSeek decision and how I split tasks across them, see my Talk-to-Build Stack.
FAQ
Is “Spud” the official name?
Internal codename. The public release name will likely be GPT-5.5.
Will it be free?
Likely Pro-tier only at launch ($20/month). Free users typically get older models.
Should I switch from Claude to GPT-5.5?
Don’t decide before testing on your own work. Both companies are leapfrogging each other every quarter. Loyalty to a brand is more expensive than $20/month for the one you don’t pick.
What about the browser feature?
If it’s the “ChatGPT browses the web while answering” model, it’s similar to features Perplexity already does well. The unlock is in how it’s integrated, not the feature itself.
Will my current ChatGPT custom GPTs work with 5.5?
Yes — OpenAI usually ports custom GPTs forward. Watch for new capabilities you can layer in once the model upgrades.
Want help applying this?
Four ways to go deeper:
- Build with Builders. Join the Talk-to-Build community to Learn how to Earn money with AI, Download our AI Skills, Advance your business, Learn to build real assets for Website Design & Shopify stores — Gen-AI images, cinematic AI videos, conversational AI office secretaries — that you can sell to SMBs that want the outcomes but don’t have time to learn the skills.
- Done-for-you. MK-Way builds AEO-ready websites and apps for design agencies and founders who want it shipped fast.
- Quick question. DM me on Instagram. I read every message.
- B2B / strategy. Connect on LinkedIn for deeper conversations about AI in design and agency work.
Last updated: May 7, 2026.